arXiv 2512.038522025-12-03cs.CV

恶劣天气下交通图像修复的频率感知Mamba

Traffic Image Restoration under Adverse Weather via Frequency-Aware Mamba

School of Communication and Information Engineering, Shanghai University（上海大学通信与信息工程学院）；College of Electronic Science, Air Force Aviation University（空军航空大学电子科学学院）；Intelligent Visual Information Perception Laboratory, Institute of Advanced Technology, Nanjing University of Posts and Telecommunications（南京邮电大学先进技术研究院智能视觉信息感知实验室）；School of Computer Science and Technology, East China Normal University（华东师范大学计算机科学与技术学院）

Liwen Pan, Longguang Wang, Guangwei Gao, Jun Wang, Jun Shi, Juncheng Li

备注：: 12pages, 13 figures, 5tables

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出FAMamba框架，通过整合频率指导与序列建模，实现恶劣天气下交通图像的有效修复。

摘要：

恶劣天气下的交通图像修复仍然是智能交通系统面临的关键挑战。现有方法主要集中在空间域建模，但忽视了频域先验。尽管新兴的Mamba架构通过块级相关性分析擅长长距离依赖建模，但其在频域特征提取方面的潜力尚未被探索。为此，我们提出了频率感知Mamba（FAMamba），一种新的框架，通过将频率指导与序列建模相结合，实现高效的图像修复。我们的架构由两个关键组件组成：（1）双分支特征提取块（DFEB），通过双向2D频率自适应扫描增强局部-全局交互，根据子带纹理分布动态调整遍历路径；（2）先验引导块（PGB），通过小波基于的高频残差学习优化纹理细节，使图像重建具有高质量和精确细节。同时，我们设计了一种新的自适应频率扫描机制（AFSM）用于Mamba架构，使Mamba能够在不同的子图中实现频域扫描，从而充分利用子图结构内固有的纹理分布特性。大量实验表明了FAMamba的效率和有效性。

英文摘要：

Traffic image restoration under adverse weather conditions remains a critical challenge for intelligent transportation systems. Existing methods primarily focus on spatial-domain modeling but neglect frequency-domain priors. Although the emerging Mamba architecture excels at long-range dependency modeling through patch-wise correlation analysis, its potential for frequency-domain feature extraction remains unexplored. To address this, we propose Frequency-Aware Mamba (FAMamba), a novel framework that integrates frequency guidance with sequence modeling for efficient image restoration. Our architecture consists of two key components: (1) a Dual-Branch Feature Extraction Block (DFEB) that enhances local-global interaction via bidirectional 2D frequency-adaptive scanning, dynamically adjusting traversal paths based on sub-band texture distributions; and (2) a Prior-Guided Block (PGB) that refines texture details through wavelet-based high-frequency residual learning, enabling high-quality image reconstruction with precise details. Meanwhile, we design a novel Adaptive Frequency Scanning Mechanism (AFSM) for the Mamba architecture, which enables the Mamba to achieve frequency-domain scanning across distinct subgraphs, thereby fully leveraging the texture distribution characteristics inherent in subgraph structures. Extensive experiments demonstrate the efficiency and effectiveness of FAMamba.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。